Live vaccines have a long history for providing efficient protection against subsequent infectious challenge. However, the mechanisms leading to protection are in many cases not well defined. Our goal is to define mechanisms for immune protection by Gram-negative bacterial vaccine strains, using novel Yersinia pestis strains as model systems. The gram-negative bacterium Yersinia pestis is the causative agent of plague. Currently there is no available licensed plague vaccine, and exploratory vaccines have variable ability to protect against pneumonic disease, the form expected after a bioterror attack. We have developed a new method for the generation of efficient vaccine strains for protection against plague and potentially other microorganisms, based upon enhancement of inherent bacterial Toll-like receptor (TLR)-4 mediated adjuvant activity. Similar to various other gram-negative bacteria, Y. pestis produces a lipopolysaccharide (LPS) with low stimulatory ability at 37:C. TLR4 is the cellular receptor for LPS via its lipid A. We generated a new Y. pestis strain expressing LpxL, an E. coli lipid A biosynthesis enzyme, and found this to produce a potent LPS at 37:C. This strain is avirulent in mice by peripheral inoculation, due to induction of antibacterial innate immune mechanisms via TLR4, a pathway also associated with strong adjuvant effects. Our results indicate that vaccination of mice with the Y. pestis LpxL strain induces full protection against both subcutaneous and intranasal challenge of mice with virulent bacteria, mimicking bubonic and pneumonic plague. Our main hypotheses are that many live bacterial vaccine strains containing LPS with increased potency are efficient vaccines, and that the increased TLR4 signaling will provide enhanced adaptive immune responses. We propose to determine mechanisms influencing the vaccine efficacy using live and killed Y. pestis producing a potent LPS, by comparing to strains without increased TLR4 stimulation, testing both in vitro and in vivo responses. Both existing and novel attenuated strains will be used. Relying on primary dendritic and T cells and genetically deficient mice, we will study TLR signaling pathways leading to dendritic cell activation in vivo and in vitro, antigen presentation and T cell subset activation. We will analyze vaccine effects against both subcutaneous and intranasal infection. The completion of these studies will provide information on the mechanism by which vaccine strains towards Gram-negative infections may act. Incorporation of TLR- stimulating adjuvant activity directly into immune-evading pathogens may constitute a novel method for attenuation and generation of vaccines.